跳到主要內容 :::

年度	111	專案性質	非實驗性質
專案類別	先導專案	研究主題	自訂
申請機構	中國文化大學	申請系所	土地資源學系
專案主持人	陳怡君	職等/職稱	助理教授
專案中文名稱	從區域永續動態發展解析污染土地對國土治理之活化策略
中文關鍵字	動態永續發展、機器學習、國土規劃
專案英文名稱	The redevelopment strategies of the pollution sites for national spatial planning from the perspective of regional dynamic sustainable development.
英文關鍵字	Dynamic Sustainable Development,Machine Learning,National Spatial Planning
執行金額	410,000元
執行期間	2022/7/1 至 2023/6/30
計畫中文摘要	褐地再利用（Brownfield Redevelopment）可視為區域發展重要策略，大部分學者常將褐地視為「污染」、「閒置」、「都市開發上具有阻礙」之土地。然而，土地具有多重生產功能，從國際諸多文獻說明現今土地利用發展變化複雜且快速，土地再利用決策若無法即時反映區域變化而調整，其污染土地管理上仍僅能投入大量污染整治經費讓土地解列，導致土地政策與鄰近區域永續發展方向仍是脫鉤。 台灣於2018年制定了永續發展目標，公布18項永續發展核心目標及對應143項子項指標。2020年頒布國土計畫法，以執行國土功能來劃分出四個使用功能分區，確保國土安全及國土永續發展，但國土復育區並沒有針對污染場址或廢棄場址的開發做規劃。土壤及地下水污染整治法是基於污染物移除與降低風險為目的制定，雖然法規提及污染整治結合健康風險及土地利用調和概念，但區域發展影響因子眾多，土地再利用產生環境衝擊與效益應更綜觀思考，包含鄰近區域發展的影響因子。綜合上述，本研究旨為探討褐地（Brownfield）地區應加速解列場址之區域重要度排序。透過統計方法的熵權法（Entropy Weight Method, EWM）機器學習（Machine Learning）的主成分分析（Principal Component Analysis, PCA）與支持向量機（Support Vector Machine, SVM），建立模型預測未來的永續發展潛力。本計畫依據空間預測分析結果，產出 SDGs 預測地圖疊合國土四大功能分區區域，其賦予土地活化後對自然環境容受力、公共設施、空間發展、土地價值等功能之路徑，產出結果可提出繪製污染土地的國土復育區必要性、可行性、迫切性等方向。此外本計畫結果可進一步探討該場址之再利用策略如何強化永續發展核心目標，以提出對永續發展之合適褐地再利用策略。
計畫英文摘要	Brownfield Redevelopment can be regarded as an important strategy for regional development. Most scholars often regard brownfields as "polluted", "idle" and "blocked in urban development". However, land has multiple production functions. Many studies show that the development and change of land use are complex and rapid. If the decision of land reuse cannot be adjusted immediately to reflect regional dynamic changes, the management of polluted sites only invests a large amount of the remediation funds to clean the pollutants but decouple from land policy and regional sustainable development. Taiwan government announced the Sustainable Development Goals(SDGs) in 2018, including 18 core goals and 143 sub-indicators to adopt localization. In 2020, the National Spatial Planning Act was announced to divide four functional zones to ensure land security and sustainable development of the national land. However, the National Spatial Planning Act never considered the polluted or abandoned sites in the implementation chapter of Land Restoration and Rehabilitation. The Soil and Groundwater Pollution Remediation Act is formulated based on the removal of pollutants and risk reduction. Although the regulations mention that the principle of site remediation combines health risks and the concept of land use reconciliation, there are many factors affecting regional development. As the result, the polluted sites reuse needs more comprehensive thinking, including the influencing factors of the regional sustainable development. The purpose of this project is to explore the priority ranking from the polluted sites to the Brownfield redevelopment area. The machine learning method is used to quantify the development sustainability of each region by Principal Component Analysis(PCA) and Support Vector Machine(SVM) predicts the sustainable development potential of the polluted sites. Based on the results of spatial prediction analysis, this project produces SDGs prediction maps overlaid with the four functional regions of the National Spatial Planning and provides the decision pathway for the necessity, feasibility, and urgency of polluted land restoration. Consequently, the context of regional sustainable development changes strengthens the core goal of sustainable development by applying the adaptive brownfield redevelopment strategies for sustainable development in this project.